¹ For consistency the names of the months and days of the Zoroastrian calendar are given throughout this article in standard Middle Persian forms, even in citations from Arabo-Persian writings or in reference to current usages: v is used rather than w because forms such as Fravardīn and Ardvahišt are well established, and the familiar Fravardīgān is kept, although this name was evidently contracted to Frōrdīyān in popular speech, at least by the late Sasanian period.

² Al-Bīrūnī, , The chronology of ancient nations, ed. and transl. by Sachau, E., 217Google Scholar.

³ See Christensen, A., ‘Some notes on Persian melody-names…’, in The Dastur Hoshang memorial volume, Bombay, 1918, 376Google Scholar, with Les types du premier homme et du premier roi dans l'histoire légendaire des Iraniens, II, Leiden, 1934, p. 153, n. 3Google Scholar.

⁴ See Burḥān-i qāṭi' s.v. Nōrōz; further Unvala, M. R., ‘A few Parsee festivals (jashans) according to an old Parsee manuscript’, in Modi, J. J. (ed.), Spiegel memorial volume, Bombay, 1908, 204–5Google Scholar.

⁵ Seervai, K. N. and Patel, B. B., Gazetteer of the Bombay Presidency, IX, Pt. II, 1899, 219Google Scholar.

⁶ This is the practice, for example, of Dastur Khodadad Shahriyar Neryosangi, now of Sharīfābād.

⁷ In Yazd and its villages the ‘Great Nō Rōz’ is now known as Habdorū (Havdorū, Habzorū), a name perhaps derived from Haft Dārūg, see Kitāb al-maḥāsin wa 'l-aḍdād, ed. van Vloten, , Leyden, 1898Google Scholar (repr. 1966), 361, transl. E. Ehrlich, ‘The celebration and gifts of the Persian New Year (Nawrūz) according to an Arabic source’, in Dr. Modi memorial volume, Bombay, 1930, 98.

⁸ op. cit., 222, 223; concerning Nō Rōz, ibid., 218. As with Nō Rōz, Sasanian melody names survive for Mihragān which appear to be connected with the two festivals, namely Mihragān ī and Mihragān ī xwurdag, see Christensen, , Dastur Hoshang memorial volume, 376Google Scholar.

⁹ See Firdausī, , Shāhnāma, ed. Vullers, , I, 25, vv. 47–55Google Scholar; al-Bīrūnī, , op. cit., 216, 233Google Scholar; al-Tha'ālibī, , Histoire des rois des Perses, ed. and transl. by Zotenberg, H., 13Google Scholar; al-Tabarī, , Annales, ed. de Goeje, , I, 180.12 ffGoogle Scholar., transl. Christensen, , Les types du premier homme…, II, 86Google Scholar.

¹⁰ Al-Bīrūnī, , op. cit., 222Google Scholar. Otherwise al-Tha'ālibī, , op. cit., 35–6Google Scholar.

¹¹ Christensen, , Les types du premier homme, II, 144Google Scholar, held that the double Nō Rōz was first evolved in Islamic times. On the duplicated Mihragāns see Taqizadeh, S. H., Old Iranian calendars, London, 1938, 44–5Google Scholar, and further Henning, W. B., ‘The murder of the Magi’, JRAS, 1944, p. 134, n. 1Google Scholar.

¹² See von Gutschmid, A., ‘Über das iranische Jahr’, Berichte der Kōniglich Sächsischen Gesellschaft der Wissenschaften, 1862Google Scholar (= KleineSchriften, III, Leipzig, 1892, 209–15Google Scholar); Markwart, J., ‘Das Naurōz, seine Geschichte und seine Bedeutung’, Dr. Modi memorial volume, Bombay, 1930, 721–3Google Scholar; Taqizadeh, , ‘ The Iranian festivals adopted by the Christians and condemned by the Jews ’, BSOAS, X, 3, 1940, 635Google Scholar (where he establishes inter alia that the Nausardēl of the Nestorians derived its name from the Great, not from the Lesser Nō Rōz). These scholars all attributed the duplication to an intercalation of the Achaemenian period. Lewy, H., ‘ Le calendrier perse ’, Orientalia, NS, X, 1–2, 1941, 1–64Google Scholar, suggested that it arose at the beginning of the Sasanian period; but against her reasoning see Bickeiman, E. J., ‘The “Zoroastrian” calendar’, Archiv Orientálni, XXXV, 2, 1967, 203Google Scholar.

¹³ Great Bundahišn, i a, 16, ed. Anklesaria, T. D., Bombay, 1908, 22.10Google Scholar, transl. B. T. Anklesaria, Bombay, 1956, 27.

¹⁴ See Christensen, , Dastur Hoshang mem. vol., p. 369, n. 2Google Scholar. Further instances can be added to those given there.

¹⁵ GBd., ed. , T.D.A., 55.9–10Google Scholar.

¹⁶ Dhabhar, B. N. (ed.), The Pahlavi Rivāyat accompanying the Dādistān ī dīnīk, Bombay, 1913, LXV1, 9 (p. 208)Google Scholar; see Bailey, H. W., Zoroastrian problems in the ninth-century books, Oxford, 1943, 138Google Scholar.

¹⁷ op. cit., 11.

¹⁸ The Hindu sāvana year of 360 lunar days and 12 months, likewise apparently originally devised for liturgical computation, also came to be used for chronological reckonings and historical events; see Barnett, L. D., The antiquities of India, London, 1913, 194–5, 203–5Google Scholar. The ancient Iranians similarly divided the lunar as well as the solar month into 30 days, see, e.g., Anklesaria, B. T. (ed.), Vichitakiha-i Zatsparam, Bombay, 1964, XXXIV, 26Google Scholar.

¹⁹ On the lack of evidence for its use before the Parthian period see Bickerman, , art. cit., 204–5Google Scholar. In the Aramaic inscription, Darius, Naqš-i Rustam b, conjecturally dated to the first half of the third century B.C. (see Henning, , Handbuch der Orientalistik, Abt. I, IV. Bd., Iranistik, 1, 24)Google Scholar, Herzfeld, E. read the words m'hy sndrm (see his Altpersische Inschriften, Berlin, 1938, 12)Google Scholar. His hand-copy shows the m of the second word as final, and the date appears too early for -nt > -nd. Even if substantiated, therefore, the reading could hardly be interpreted as ‘in the month Sandārm[ad]’, i.e. Spendārmad. In fact no other scholar has been able to read these words, see the remarks of Cameron, G. G., Persepolis treasury tablets, Chicago, 1948, 29aGoogle Scholar.

²⁰ See D'yakonov, I. M. and Livshits, V. A., ‘Novye nakhodki dokumentov v staroi Nisa’, Peredneaziatskii sbornik, II, Moscow, 1960, 135–57Google Scholar(English summary, 169–73). (I am indebted to Dr. Livshits for his kindness in entering into correspondence with me about the unpublished material for the Nisa calendar.)

²¹ See Welles, C. B., Royal correspondence in the Hellenistic period, New Haven, 1934, 299 fGoogle Scholar.

²² See D'yakonov, and Livshits, , op. cit., 143–1 with n. 28Google Scholar; Dokumenty iz Nisy I v. do N.E., Moscow, 1960, 20, 113Google Scholar; Bickerman, , ‘The Parthian ostracon No. 1760 from Nisa’, Bibliotkeca Orientalis XXIII, 1–2, 1966, 15–17Google Scholar; Chaumont, M.-L., ‘Les ostraca de Nisa, nouvelle contribution à l'histoire des Arsacides’, JA, CCLVI, 3, 1968, 17–19, 35Google Scholar.

²³ See al-Bīrūnī, , op. cit., 11Google Scholar.

²⁴ See Henning, , art. cit., 29Google Scholar (for Avrōmān); and ‘The monuments and inscriptions of Tang-i Sarvak’, Asia Major, NS, II, 2, 1952, 176Google Scholar(for the Ardabān inscription).

²⁵ Non-Iranians continued to use the Babylonian lunisolar calendar, see Bickerman, , Archiv Orientální, XXXV, 2, 1967, 205Google Scholar, and ‘Time-reckoning in Iran’, in Yarshater, E. (ed.), The Seleucid, Parthian and Sassanian periods (Cambridge History of Iran, III)Google Scholar, in press. (I am much indebted to Professor Bickerman for letting me read this chapter in draft and for subjecting a draft of the present article to trenchant and valuable criticism.)

²⁶ See Andreas, F. C. and Henning, W. B., Mitteliranische Manichaica aus Chinesisch-Turkeslan, I (SPAW, Phil.-hist. KI., 1932, 10), 190.16–19, cf. 191.3–5Google Scholar.

²⁷ Henning, , Ein manichäisches Henochbuch (SPAW, Phil.-hist. Kl., 1934, 5), 33Google Scholar, attributed the text to between A.D. 235 and 238, i.e. to the reign of Ardašīr. Lewy, H., art. cit., 36–7Google Scholar, assigned it to c. A.D. 244, i.e. the beginning of the reign of his son Šābuhr.

²⁸ Al-Bīrūnī, , op. cit., 32Google Scholar. (See Lewy, , art. cit., 41Google Scholar, and contra, Bickerman, , Archiv Orientální, XXXV, 2, 1967, 203Google Scholar.)

²⁹ See Tansarnāma (The letter of Tansar), ed. Minovi, M., Tehran, 1932, 10–12Google Scholar, transl. M. Boyce, Rome, 1968, 36–7, and intro., p. 22, n. 2.

³⁰ That Fravardīn was the first calendar-month at the beginning of the Sasanian period is established by the inscription of Šābuhr I at Bīshāpūr, see Ghirshman, R., ‘Inscription du monument de Châpour I^er’, Revue des Arts Asiatiques, X, 3, 1937, 123–9Google Scholar.

³¹ See Geiger, W., Civilisation of the Eastern Iranians, transl. Sanjana, D. P., London, 1885, I, p. 148, n. 2, 149Google Scholar; Taqizadeh, , Old Iranian calendars, p. 10Google Scholar with n. 2, 46. With regard to the Achaemenian festival of the Magophonia, , Henning, (JRAS, 1944, p. 134, n. 1)Google Scholar pointed out that Markwart was not justified in calling this a five-day festival, since Herodotus states that it occupied only a single day.

³² Saddar Bundahiš, lii, 2 (text in Dhabhar, B. N. (ed.), Saddar Naṣr and Saddar Bundehesh, Bombay, 1909, 125–6Google Scholar; transl., Dhabhar, , The Rivāyat of Hormazyar Framarz, 542–3Google Scholar).

³³ Dīnkard, V, 29 (ed. D. P. Sanjana, x).

³⁴ GBd., ia, 21, ed. , T.D.A., 24.4–5Google Scholar (rōz ī truftag/duzīdag). These names were rendered in Arabic as al-masrūqa and al-mustaraqa, see al-Bīrūnī, , op. cit., 43Google Scholar.

³⁵ op. cit., 224.

³⁶ The 10-day Fravardīgān festival is referred to in Pahl. Vd., viii, 22; Nīrangistān, ed. Sanjana, D. P., fol. 52r., 15 ffGoogle Scholar., transl. Bulsara, S. J., Aērpatastān and Nīrangastān, 111 ff.Google Scholar; Dk., VIII, 6.11 (ed. Sanjana, , XV)Google Scholar, as well as in later works. In VZ, XXXV, 19 ff. (ed. , B.T.A., 155 ffGoogle Scholar.) the rituals and events of the end of the world-year are amplified to last 10 days. For the actual observance of the 10-day festival in A.D. 565 see Doblhofer, E. (tr.), Byzantinische Diplomaten und ōstliche Barbaren, aus den Excerpta de legationibus des Konstantinos Porphyrogennetos ausgewählte Abschnitte des Priskos und Menander Protektor, Graz, 1955, 122Google Scholar. The expression ‘the farvardiyān days’ is still used of the festival in the Persian Rivāyats.

³⁷ Information from Ervad Dr. Firoze M. Kotwal.

³⁸ See Modi, J. J., The religious ceremonies and customs of the Parsees, second ed., Bombay, 1937, 440Google Scholar. In the Persian Rivāyats there is a tendency to regard the ‘Farvardiyān days’ as extending up to 6 Fravardin (see Unvala, M. K., Dārāb Hormazyār's Rivāyat, Bombay, 1922, I, 506.13Google Scholar, transl. Dhabhar, B. N., The Persian Rivayats of Hormazyar Framarz, Bombay, 1932, 337)Google Scholar, although they are still numbered as ten.

³⁹ Taking vīsāδa to mean ‘to their homes’, with Haug, Justi, Darmesteter, and others. This translation has in general been adopted despite th e assumption that vīsāδa was an ablative; but Henning interpreted this word as having the directive suffix -da (cf. vaēsmənda, GIF, I, § 304.II.10) with anomalous preceding ā.

⁴⁰ The data for these northern calendars were conveniently brought together by Gray, L. H., ‘On certain Persian and Armenian month-names as influenced by th e Avesta calendar’, JAOS, XXVIII, 2, 1907, 331–44Google Scholar. For the older Khwar. material discovered since see Livshits, , ‘The Khwarezmian calendar and the eras of ancient Chorasmia’, Ada Antiqua Acad. Scient. Hungaricae, XVI, 1–4, 1968, 433—46Google Scholar. On the difference of five days between the Sogd. and Persian calendars see al-Bīrūnī, , op. cit., 46, 233Google Scholar; and on the Armenian feast-days Taqizadeh, , BSOAS, X, 3, 1940, 639 ffGoogle Scholar. Henning has pointed out that the evidence from Mt. Mugh suggests that the Sogdians did not follow the Sasanians in naming the epagomenae, but simply called them after the first five days of the month, see his ‘A Sogdian god’, BSOAS, XXVIII, 2, 1965, p. 251, n. 58Google Scholar.

⁴¹ See Hertel, J., ‘Die awestischen Jahreszeitenfeste. Afrīnagān 3’, Berichte über die Verhandlungen der Sächsischen Akademie der Wissenschaften, Phil.-hist. KL, XI, 2, 1934, 22Google Scholar; Nadershah, J. D., ‘The Zoroastrian months and years…’, The K. R. Cama memorial volume, Bombay, 1900, 252–4Google Scholar.

⁴² VZ, XX, 1, xxi, 1, ed. , B.T.A., 77, 78Google Scholar. Anklesaria (text and intro., p. xciii) is undoubtedly right in preferring the reading ZK-š 5 YWM to the variant ZK 40 5 YWM. (Otherwise Molé, M., Culte, mythe et cosmologie dans l'Iran ancien, Paris, 1963, 317Google Scholar.)

⁴³ op. cit., 224.

⁴⁴ Nīr. fol. 53r., 4 (Bulsara, , 113)Google Scholar.

⁴⁵ Nīr. fol. 52v., 6 (Bulsara, , 112)Google Scholar.

⁴⁶ See Boyce, , ‘Festivals’, in Yarshater, E. (ed.), The Seleucid, Parthian and Sassanian periods (Cambridge History of Iran, III)Google Scholar, in press.

⁴⁷ op. cit., 219.

⁴⁸ op. cit., 217.

⁴⁹ Saddar Bd. 1, 3–19 (Dhabhar, (ed.), 122Google Scholar, (tr.), 541).

⁵⁰ See Dhabhar, (ed.), Zand-i Khūrtak Avistāk, Bombay, 1927, 152.4 f., 372–3Google Scholar; transl. (with MS variants), Bombay, 1963, 293–4; earlier transl. by Darmesteter, , ZA, III, 181–2Google Scholar.

⁵¹ See GBd., ia, 16 ff. (ed. , T.D.A., 22.8–11)Google Scholar.

⁵² GBd., v b, 4, ed. , T.D.A., 56.5–9Google Scholar; Nyberg, H. S., Texte zum mazdayasnischen Kalender, Uppsala, 1934, 25Google Scholar(with slightly different translation).

⁵³ See GBd., XXV, 1, ed. , T.D.A., 157.12–14Google Scholar (360 days), against the Indian Bd., ed. Justi, F., Leipzig, 1868, 59Google Scholar (365 days); see Nyberg, , op. cit., 10Google Scholar.

⁵⁴ VZ, xxxiv, 49, ed. , B.T.A., 148Google Scholar, intro., cxxii.

⁵⁵ See Nöldeke's, transl. of al-Tabarī, Geschichte der Perser und Araber, Excursus 1, 400–34Google Scholar.

⁵⁶ op. cit., 44.

⁵⁷ See Boyce, , ‘Rapithwin, Nō Rūz and the feast of Sade’, in Heesterman, J. C., Schokker, G. H., and Subramoniam, V. I. (ed.), Pratidānam… studies presented to F. B. J. Kuiper, The Hague, 1968, p. 202, n. 8Google Scholar.

⁵⁸ See Taqizadeh apud Minorsky, V., ‘Vīs u Rāmīn (II)’, BSOAS, XII, 3, 1947, 35Google Scholar.

⁵⁹ Al-Bīrūnī, , op. cit., 45Google Scholar (Yazdegerd); Qānūn-i Mas'ūdī, I, Hyderabad, 1954, 132Google Scholar (Pērōz); see Taqizadeh, , Calendars, 37Google Scholar; Bickerman, , art. cit., p. 202Google Scholar with n. 23. The reform has also been assigned to the reigns of Xusrau I (531–79) and even Yazdegerd III (632–51); but there is evidence to show that these attributions are too late (see further below). The melody called the Nō Rōz ī Kay Kavād (see Christensen, , Dastur Hoshang mem. vol., 375Google Scholar) was presumably composed in honour of the new Nō Rōz.

⁶⁰ See Nyberg, op. oit., 60Google Scholar.

⁶¹ See Hoffmann, G., Auszüge aus syrischen Akten persischer Märtyrer (Abh. für die Kunde des Morgenlandes, VII, 3), 1880, 79Google Scholar.

⁶² Pahl. Riv. Dd., i, 2; Nyberg, , op. cit., 44Google Scholar.

⁶³ Pahl. Riv. Dd., i, 4; cf. Vizīrīhā ī dēn ī weh ī Mazdēsnān, ed. Jamasp-Asa, K. J., in Boyoe, M. and Gershevitch, I. (ed.), W. B. Henning memorial volume, London, 1970, 206Google Scholar. In the Khwar. calendar, which was not affected by the second Sasanian reform, the day most sacred to the departed remained Rōz Fravardīn of Māh Fravardīn, cf. the Toq-qal'a ossuary no. 25 (see Henning, , ‘The Choresmian documents’, Asia Major, NS, XI, 2, 1965, 179)Google Scholar.

⁶⁴ See Mādigān ī hazār dādistān, ed. Modi, J. J., Bombay, 1901, 35.13–14Google Scholar.

⁶⁵ op. cit., 224.

⁶⁶ Prairies d'or, § 1298, ed. Pellat, Ch., Paris, 1965, 495Google Scholar.

⁶⁷ See Unvala, , Spiegel mem. vol., 208Google Scholar; and cf. al-Bīrūnī, , op. cit., 225Google Scholar.

⁶⁸ See Boyce, in Heesterman, and others (ed.), Pratidānam, 201–5Google Scholar.

⁶⁹ To-day, conversely, the Zoroastrians celebrate the Nō Rōz ī Jamšēdl as a secular feast in spring, and the religious Nō Rōz on I Fravardīn, which is once more in August. On the secular and religious Nō Rōz in Sasanian times see also Kuka, M. N., ‘Principal Persian festivals in th e days of Naosherwan’, Sir Jamsetjee Jejeebhoy Madressa jubilee volume, Bombay, 1914, 11–14Google Scholar.

⁷⁰ Dk.M, 683.7–8; Nyberg, , op. oit., 6Google Scholar.

⁷¹ See Hyde, Th., Historia religionis veterum Persarum, Oxford, 1700, 238Google Scholar. (Otherwise interpreted by Lewy, , art. cit., p. 42Google Scholar, n. 1.)

⁷² VZ, XXV, 5 (ed. , B.T.A., 91–2)Google Scholar; see Taqizadeh, , ‘Some chronological data relating to the Sasanian period’, BS08, IX, 3, 1937, 132Google Scholar(with a different interpretation).

⁷³ See Sachau's ed., 45.

⁷⁴ See Bickerman, , art. cit., 199Google Scholar.

⁷⁵ Georgios Chrysokokkes, writing in A.D. 1346, see Gray, L. H., ‘Medieval Greek references to the Avestan calendar’, Avesta, Pahlavi and ancient Persian studies in honour of…P. B. Sanjana, I, Strassburg, 1904, 170Google Scholar. A generation later, Isaakos Argyros tells of only the one Zoroastrian calendar (see Gray, ibid., 173–4). For Dk., III, see Nyberg, , op. cit., 30–9Google Scholar, with his remarks, 80—6.

⁷⁶ GBd., i, 50, ed. , T.D.A., 12.15–13.2Google Scholar.

⁷⁷ Al-Bīrūnī, , op. cit., 55Google Scholar.

⁷⁸ See Unvala, M. B., ‘Two Persian passages about the kabiseh (intercalation)’, K. R. Cama mem. vol., Bombay, 1900, 235–6Google Scholar.

₇₉ Dk.M, 402.20–403.11; Nyberg, , op. cit., 32–4Google Scholar (with some differences in translation; see his edition for Pahl. words marked with an asterisk in the text reproduced here). Mihragān is evidently kept in the argument because of its popularity. The position of this feast, as we have seen, was not altered by the second reform.

⁸⁰ This appears to be an allusion to non-conformity over the position of Nō Rōz, and over the second reform, on the part of the northern kingdoms.

⁸¹ Jamasp-Asana, J. M. (ed.), The Pahlavi texts, II, Bombay, 1913, 102–7Google Scholar; Blochet, E. (tr.), Revue archéologiqne, Paris, 1895, 17–22Google Scholar; Jamasp-Asana, K. J. (tr.), K. R. Cama mem. vol., Bombay, 1900, 122–9Google Scholar; Markwart, J. (tr.), Dr. Modi mem. vol., Bombay, 1930, 742–65Google Scholar.

⁸² Text, § 11.

⁸³ cf. Saddar Bd., lii, 2. Although 6 Fravardīn was thus celebrated as the Great Nō Rōz, it was still invested with some observances proper to Hordād, the Amešaspand who protects water. See Kitāb al-maḥāsin, ed. van Vloten, , 362Google Scholar, transl. Ehrlich, , Modi mem. vol., 99Google Scholar; Taqizadeh, , BSOAS, X, 3, 1940, p. 636 with n. 2Google Scholar. The importance of Rōz Hordād in Zoroaster's own life is stressed in Vizīrkerd ī dīnīg, ed. Sanjana, P., Bombay, 1848Google Scholar, i, §§ 6, 17, transl. Molé, M., La légende de Zoroastre selon lea textes pehlevis, Paris, 1967, 125, 131Google Scholar.

⁸⁴ Vīs u Rāmīn, ed. Minovi, M., Tehran, 1935, 148Google Scholar; see Minorsky, V., ‘Vīs u Rāmīn, a Parthian romance’, BSOAS, XI, 4, 1946, p. 747Google Scholar, n. 2.

⁸⁵ Vīs u Rāmīn, ed. Minovi, , 44Google Scholar; see Taqizadeh apud Minorsky, , BSOAS, XII, 3, 1947, 34–5Google Scholar. On the Persian custom of marrying at the spring equinox see Strabo, , Geographia, XVII, 733Google Scholar.

⁸⁶ On this festival see Boyce, , ‘Ātaš-zōhr and Āb-zōhr’, JRAS, 1966, 107Google Scholar. With regard to the gahāmbārs, the first da y came in time to be the one chiefly celebrated, see Sheriyar, Khudayar Dastur, ‘The celebration of the gahāmbār in Persia’, Sir Jamsetjee Jejeebhoy Madressa jubilee volume, Bombay, 1914, 302Google Scholar; Boyce, , BSOAS, XXXI, 2, 1968, 277Google Scholar.

⁸⁷ op. cit., 220. Tiragān is not mentioned in the Shāhnāma; and it seems likely that the epic link, with its pun on the word for arrow, tīr, was first made in the later Sasanian period. In the third century A.D. the older form of the word, tigr, was still in use.

⁸⁸ Salemann, C., Manichaeische Studien (Mém. de l'Ac. Impériale des Sciences de St. -Péters-bourg, VIII^e Sér. VIII, 10), 1908, 8Google Scholar; text restored by Henning, , JRAS, 1944, p. 134, n. 1Google Scholar.

⁸⁹ See Taqizadeh, , ‘An ancient Persian practice preserved by a non-Iranian people’, BSOS, IX, 3, 1938, p. 608Google Scholar with n. 3. The great feasts of Mihragān and Tīragān were not the only nonobligatory ones to be duplicated, for al-Bīrūnī records that Sada was also celebrated again after a five-day interval, the second celebration being as usual the major one; see his Book of instruction in the elements of the art of astrology (Kitāb al-tafhīm li awā'il ṣinā'at al-tanjīm), ed. and transl. by Wright, R. R., London, 1934, 310Google Scholar.

⁹⁰ For the festival among the Parsis see Modi, J. J., The religious ceremonies and customs of the Parsis, second ed., 435Google Scholar; Seervai, and Patel, , Gazetteer of the Bombay Presidency, IX, Pt. II, 217Google Scholar; for the Irani one as formerly observed see Unvala, , Darab Hormazyar's Rivāyat, I, 526.8—19, 527.1–6Google Scholar; Dhabhar, transl., 341–2; further Unvala, , Spiegel mem. vol., 206Google Scholar.

⁹¹ op. cit., 209.

⁹² See Taqizadeh, , BSOS, IX, 3, 1938, 604–6Google Scholar; IX, 4, 1939, 917–18.

⁹³ cf. above, p. 525, ga'āmbār-i panjīvak. The month-names for the gahāmbārs are now usually taken from the jalālī calendar.

⁹⁴ cf. the old use, in VZ, of the term ‘Spring festival’ for Maiδyōizarəmaya, see above, p. 524.

⁹⁵ The 10-day festival was kept by the Zoroastrians of Yazd down to the early decades of the present century, and is still remembered by older members of the community there. The observance must have been established at the time of the calendar change yet in the Rivāyat of Shapur, Kamdin, dated A.D. 1558 (for references see above, p. 535, n. 90)Google Scholar the festival is still described as lasting one day only. Presumably the 10-day feast was a popular development carried out in opposition to scholar-priests, in whose families the older tradition, enshrined in written texts, may have remained known for generations.

⁹⁶ See Karaka, D. F., The history of the Parsis, I, London, 1884, 151Google Scholar.

⁹⁷ See Khareghat, M. P., ‘The Daryāī Nōrōz’, Dr. Modi mem. vol., 118–30Google Scholar.

⁹⁸ Henning insisted that the ‘258 years before Alexander’ should be regarded as a genuine date, and not a reconstructed one. For his discussion of it see his Zoroaster, politician or witch-doctor?, Oxford, 1951, 36 ffGoogle Scholar. A date of 660 B.C. for Zoroaster's birth is indicated by the legendary chronology, see E. W. West, SBE, XLVII, intro., § 55.

⁹⁹ The fact that their names survive in YAv. forms (having undergone presumably the regular changes of a living language) is not incompatible with this supposition. The gahāmbārs are the only Zoroastrian feasts whose foundation is traditionally ascribed to the prophet, see al-Bīrūnī, , op. cit., 219Google Scholar.

¹⁰⁰ On the closeness of the link between Aša and Rapithwin see, e.g., Boyce, in Heesterman, and others (ed.), Pratidānam, 203–4Google Scholar; and on the other Amešaspands and the gahāmbārs see ‘Zoroaster the priest’, BSOAS, XXXIII, 3, 1970, p. 27Google Scholar with n. 32.

¹⁰¹ Note the current Irani name for the Greater Nō Rōz (see above, p. 513, n. 7). For instances of sevenfold offerings and ceremonies at Nō Rōz see Kitāb al-maḥāsin, ed. van Vloten, , 361Google Scholar, transl. Ehrlich, 98 (where moreover the Amešaspands are evidently represented in the dawn-offerings set before the king); Unvala, , Dārāb Hormazyār's Rivāyat, I, 516.18 fGoogle Scholar., transl. Dhabhar, 339. The custom of Persian Muslims of setting out the haft sīn is not practised by Zoroastrians, but it perpetuates the link between Nō Rōz and the number seven.